Seepage Controll Measures In Geotechnical Engineering

Introduction
Seepage flow through soils, rocks and geotechnical structures has a great influence on their
stabilities and performances, and seepage control is a critical technological issue in engineering
practices. The physical mechanisms associated with various engineering measures for seepage
control are investigated from a new perspective within the framework of continuum mechanics;
and an equation-based classification of seepage control mechanisms is proposed according to
their roles in the mathematical models for seepage flow, including control mechanisms by
coupled processes, initial states, boundary conditions and hydraulic properties. The effects of
each mechanism on seepage control are illustrated with examples in hydroelectric engineering
and radioactive waste disposal, and hence the reasonability of classification is demonstrated.
Advice on performance assessment and optimization design of the seepage control systems in
geotechnical engineering is provided, and the suggested procedure would serve as a useful
guidance for cost-effective control of seepage flow in various engineering practices.

Seepage control by coupled processes

To illustrate the seepage control mechanism by coupled processes, here we focus on the
moisture transfer in an engineered barrier system of nuclear waste repository. As proved by
both in-situ experiments and theoretical predications [24], [25], the excavation of tunnels in a
hard rock may induce an increase in permeability up to 3 orders of magnitude in the disturbed
zone. Then, as compacted expansive clay blocks are placed in the surrounding space of the host
rock, deformations of the barrier materials resulted from thermal expansion and moisture
swelling are constrained, hence impact the water flow by changes in storage terms (porosity,
volumetric strain, and density) and hydraulic properties (permeability and water retention
curves). In addition, the heat generated by the radioactive decay of the waste may have an
impact on the water flow by the thermo-osmosis effect of the clay materials in the engineered
barriers. By choosing clay materials with different swelling and thermal coupling properties,
altering these properties in the manufacturing process, or restraining the significant excavation-
induced increase in permeability in the surrounding rocks with some well-controlled tunneling
techniques, the water flow process may be controlled to some extent.
According to the modeling results, it has been found that the thermo-osmosis effect takes
negligible effect on the moisture transfer in the bentonite sample, and the evolutions of relative
humidity measured by 7 capacitive sensors installed at different heights of the sample can be
best fitted by assuming isotropic thermal coupling with a thermal coupling coefficient kT = 0
m2/(s·K) in Eq.(2). The corresponding evolutions of the degree of saturation Sr at various
heights of the sample are plotted in Fig.1(a). However, if the thermal coupling coefficient is
assumed to vary linearly with the degree of saturation, i.e. kT = kT0Sr, by applying some
control mechanisms in the manufacturing procedure of the sample, with the value in saturated
state kT0 = 2×10−12 m2/(s·K) (a value comparable to that adopted for the FEBEX bentonite
[17]), the degree of saturation will evolve in the manner shown in Fig.1(b). Comparison of
Figs.1(a), (b) shows very different distribution and evolution patterns of water content in the
sample, only due to the thermo-osmosis effect by moving water in the negative direction of
thermal gradient. Obviously, the stronger the thermo-osmosis effect is, the longer the period of
time will be needed for full re-saturation of the sample, which is an important measure for
designing a nuclear waste repository

Seepage control by initial states

As stated before, the initial distribution of water content (or equivalently, suction in soil
mechanics) will lead to very different wetting processes in a low-permeability engineered
barrier. In dam engineering, the initial distribution of pore water head in the strata of the dam
site also plays an important role in the near-term evolution of seepage field, indicating the
significance for site investigation and characterization about hydrological conditions and the
possibility in controlling the seepage flow by changing its initial condition.

Seepage control by boundary conditions

As stated before, the seepage control mechanism by boundary conditions is one of the most
effective control mechanisms in engineering practices. Without loss of generality, here we
consider the possible boundary conditions of the drainage holes in unsteady seepage flow
condition [9], as illustrated in Fig.3. The first type is the water head boundary condition, as
depicted in Fig.3(a). The drainage holes deployed in a rock foundation generally possess this
type of boundary condition. The prescribed water head is usually determined by the floor
elevation of the drainage tunnel connected with the holes. The second type of boundary
condition is the Signorini’s type, such as the vertical drainage holes deployed between two
horizontal drainage tunnels in a concrete dam, where the drainage flow is always discharged
into the lower drainage tunnel. As shown in Fig.3(b), on section AB of the drainage hole, the
boundary condition satisfies ϕ < z and qn = 0, while on section BC, it satisfies ϕ = z and qn ≤ 0.
Seepage control by hydraulic properties
The seepage control mechanism by hydraulic properties is another one that has been effectively
and widely used in engineering practices. The grouting curtain in rock foundations is first
examined here for illustration of the seepage control effects. According to the “Design
specification for concrete gravity dams” (SL 319-2005) and the “Design specification for
concrete arch dams” (DL/T 5346-2006) in China, the permeability of a grouting curtain in dam
foundations must satisfy the following criteria: q = 1–3 Lu for dam height over 100 m; q = 3–5
Lu for dam height between 50 and 100 m; and q = 5 Lu for dam height below 50 m, where q (in
Lu or L/(min·MPa·m)) is defined as the flow rate of water per unit water pressure injected into
a unit section of borehole at the third (or maximum) pressure step of packer testing after steady-
state condition is achieved. As a result, the permeability of the grouted zone is well controlled
and an impervious system of integrity is created, hence the undesirable effects of concentrated
channel flow through conductive faults, weak zones, joints or fractures could be eliminated.

Optimization design of seepage control systems

Procedure for optimization design
With the above equation-based classification of seepage control mechanisms and the control
effects of various engineering measures, advice on optimization design of a seepage control
system in geotechnical engineering is provided in this section. The suggested flowchart is
plotted in Fig.6, which at least has been partly followed in engineering practices The procedure
is described as follows:
Site investigation of geological and hydrological condition
 Flowchart of optimization design of a seepage control system.

As the first step of optimization design of a seepage control system, comprehensive

investigation of geological and hydrogeological conditions in the engineering site is critical for
a clear understanding of the initial state of the seepage field, the conducting capability of the
media, the resistance of the materials to seepage failure and the in-situ stress field and
geothermal conditions of the project. At this step, the degree of fracture development, the
potential concentrated flow channels and the correlation between the permeability and the in-
situ stress should be particularly identified.

Identification of main factors that influence the seepage flow

behaviors
The main factors that may have an influence on the seepage flow behaviors should be identified
according to the hydrological nature of the site. These factors may include, but are not limited
to, the initial distribution of water content or pore water pressure in the field, the coupling
effects of thermo-mechanical processes on the seepage flow, the hydraulic properties of the
media, the boundary conditions of the problem, the climate condition in the area, the preferable
engineering measures for seepage control, the possible defects in the seepage control system,
etc‥ Sensitivities of the factors to the seepage flow should also be preliminarily assessed based
on the existing knowledge and experiences that have been accumulated..

Sensitivity analysis of the seepage control mechanisms

The above findings lead to choices of mathematical models most suitable for describing the
seepage flow behaviors in the domains of interest. The possible models may include the
saturated/unsaturated model, the non-steady state model, the steady state model, or even the
discrete fracture network flow model that is out of the scope of this paper. Then the suitability,
applicability, reliability and sensitivity of the above-mentioned four mechanisms for seepage
control in the project should be analyzed based on the mathematical models selected, together
with the initial and boundary conditions of the problem as well as the physical properties of the
materials. At this step, theoretical conceptualization and analysis need to be performed, but a
predictive numerical analysis is a more powerful tool, even though complicated calculations are
involved.

Schematic design of the seepage control system

Designing the seepage control layouts should then be performed according to the mechanisms
that may produce the most effective control of the seepage flow and the corresponding
measures that are available in engineering practices. A number of design schemes should be
proposed, with different combinations of available measures and different layout parameters.
For example, the layout parameters for a drainage hole array may include the rows, spacing,
lengths and diameters of the hole. For a clay core in soil dams, the layout parameters may be the
size of the core, the clay materials that are available in the local site, the granular composition,
the compacted density and the optimum water content for compaction.

References
[1]
Fipps G., Skaggs R.W., Nieber J.L.
Drains as a boundary condition in finite elements
Water Resources Research, 22 (11) (1986), pp. 689-707
View Record in Scopus
[2]
Xu Jiahai
Seepage control of hydraulic structures on sand/gravel foundations
Chinese Journal of Geotechnical Engineering, 8 (2) (1986), pp. 96-106
(in Chinese)
[3]
Cao Dunlu, Xu Zhongsheng, Wu Songgui
Seepage control of weakened rock foundation of Gezhouba—Erjiang sluice
Journal of Hydrodynamics, 2 (2) (1987), pp. 25-33
(Ser. A), (in Chinese)